Electricity Maps calculates the carbon intensity of electricity consumption to optimize usage at scale

If you’re an electricity enthusiast, chances are you’ve already spent several hours watching Electricity Maps and its fascinating export flow animations. This open source data visualization project has been around since 2016. But companies like Google and Samsung are increasingly relying on this rich data set to achieve their sustainability goals and empower their own users.

Electricity Maps currently employs 20 people and the company has been profitable for several years. But TechCrunch has exclusively learned that it recently raised a $5.4 million (€5 million) funding round from Transition and Revent, to move up a gear by investing in the product and business. Electricity Maps is now both a data visualization tool and an enterprise API for data-driven decarbonization – and one wouldn’t work without the other.

Founded by Olivier Corradi, a French and Danish entrepreneur and data scientist who previously worked for voice assistant startup Snips, Electricity Maps was born from a desire to learn more about electricity production and consumption. “There have been articles in the press saying that Denmark ran 100% on renewable energy for a day. So I was wondering if that was true and how long it lasted – how many hours in the day,” Corradi told TechCrunch.

Image credits: Electricity cards

Electricity Maps aggregates real-time data on electricity production in more than 50 countries around the world. While the company uses open data sources, employees and community contributors have had to create dozens of parsers to standardize this data.

Renewable energies depend on climatic conditions (for wind and solar energy in particular), the mix of energy sources is constantly evolving. This means that the carbon intensity of the energy produced also changes with the time of day and current conditions. And as you might have guessed, since electricity works much like tap water circulating through a vast network of tubes, the CO2 emissions linked to the electricity you consume can vary considerably.

Calculating CO2 emissions can be complicated because there are also many cross-border flows of electricity: some countries produce more electricity than they actually consume while others have electricity needs greater than their capacity of production. Electricity Maps has designed its own flow tracing model to understand which power plant contributes to the electricity you currently consume.

“We’re going to process all of that in our system with what we call a flow tracing algorithm, which allows us to tell you where the electricity is coming from based on where you are. It can be produced locally, but it can also come from Germany. But Germany imports from Poland, so it could potentially come from Poland, etc. So we have to do some modeling,” Corradi said.

The company also stores historical data and uses machine learning algorithms to provide 24-hour forecasts. Just as weather APIs are now a significant industry, electricity forecasting APIs could become essential business tools in the future.

Olivier Corradi, founder and CEO of Electricity Maps (Image credits: Electricity cards)

From carbon intensity to charge transfer

The reason the open source data visualization project is a core part of the company is that Electricity Maps wants to achieve as much consensus as possible. Calculating life cycle emissions from electricity generation requires peer-reviewed studies.

The company shares all its sources of emission factors. This also means that these calculation methods can evolve over time, as researchers publish new studies giving more precise results. The community can discuss and submit potential changes that will be reflected in Electricity Maps data.

As for the business part, being able to know the carbon intensity of the electricity available in a specific location and at a specific time can be a kind of superpower.

“With all the renewable energy installation targets, we are going to find ourselves in a world where the amount of intermittent renewable energy triples by 2030,” Corradi said.

“The good thing is that this request is flexible because they will be electric cars and you can change the time they charge. It will be AI training and you can choose when you are going to train these big models,” he added.

Image credits: Electricity cards

Google, one of Electricity Maps’ largest customers, has partnered with the European startup to calculate the carbon intensity of the electricity powering its data centers.

For certain tasks, such as indexing the web or training a new AI model, Google may use data from Electricity Maps for charge transfer. When it’s windier, it’s time to start additional servers. Or, when it gets dark in the United States, Google can move some compute-intensive operations to European data centers.

But Google’s own customers will also benefit from the company’s partnership with Electricity Maps. With the EU Sustainability Reporting Directive, many companies will soon be required to publish carbon accounting reports. As many businesses rely on Google Cloud for their hosting needs, they will need data to calculate their scope 3 emissions. Using historical data from Electricity Maps, this carbon data will be more accurate.

Samsung, another Electricity Maps customer, uses the startup’s data to show users the power consumption and carbon footprint of their Samsung devices. In this case, it’s more about educating users.

But Electricity Maps is much more than an educational tool. It can serve as the layer of information that decides whether it’s the right time to turn on millions of electrical devices – or, at least, whether it’s the right time to have as little impact on the planet as possible.

Why does a profitable business grow? The cash injection aims to increase Electricity Maps’ impact by harnessing gas so it can meet growing demand for smarter climate tools, according to Corradi. “The reason we raised this issue is simply that the amount of renewable energy that is currently in the system is starting to be so significant that you are starting to have the ability to enable the flexibility of the devices that you have at home or in the industrial sector. »